{"ID":2855115,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.13364","arxiv_id":"2510.13364","title":"Language as a Label: Zero-Shot Multimodal Classification of Everyday Postures under Data Scarcity","abstract":"Recent Vision-Language Models (VLMs) enable zero-shot classification by aligning images and text in a shared space, a promising approach for data-scarce conditions. However, the influence of prompt design on recognizing visually similar categories, such as human postures, is not well understood. This study investigates how prompt specificity affects the zero-shot classification of sitting, standing, and walking/running on a small, 285-image COCO-derived dataset. A suite of modern VLMs, including OpenCLIP, MetaCLIP 2, and SigLip, were evaluated using a three-tiered prompt design that systematically increases linguistic detail. Our findings reveal a compelling, counter-intuitive trend: for the highest-performing models (MetaCLIP 2 and OpenCLIP), the simplest, most basic prompts consistently achieve the best results. Adding descriptive detail significantly degrades performance for instance, MetaCLIP 2's multi-class accuracy drops from 68.8\\% to 55.1\\% a phenomenon we term \"prompt overfitting\". Conversely, the lower-performing SigLip model shows improved classification on ambiguous classes when given more descriptive, body-cue-based prompts.","short_abstract":"Recent Vision-Language Models (VLMs) enable zero-shot classification by aligning images and text in a shared space, a promising approach for data-scarce conditions. However, the influence of prompt design on recognizing visually similar categories, such as human postures, is not well understood. This study investigates...","url_abs":"https://arxiv.org/abs/2510.13364","url_pdf":"https://arxiv.org/pdf/2510.13364v1","authors":"[\"MingZe Tang\",\"Jubal Chandy Jacob\"]","published":"2025-10-15T09:53:46Z","proceeding":"cs.CV","tasks":"[\"cs.CV\",\"cs.AI\"]","methods":"[\"Language Model\"]","has_code":false}